Ventures Edge Predictions for 2026

Every year, we try to spot the “inflection point.” A few weeks into 2026, several shifts already feel especially relevant. AI is moving out of the chatbox and into products, machines, and infrastructure. Energy and compute are becoming strategic constraints, not background utilities. And culture, capital, and geopolitics are increasingly shaping what gets built and where.

To make sense of it all, the Ventures Edge team has pooled our distinct perspectives. Below, Colin, Michael, and Robert break down their predictions for 2026.

Michael’s Predictions

For years, "Edge Compute" was a buzzword for putting servers in closets. In 2026, the edge is no longer stationary, it’s moving, flying, and powering itself. We are entering the era of the Hard Edge, where the most critical AI breakthroughs aren't happening in LLM chat boxes, but in the kinetic world of robotics and decentralized power.

1. Kinetic Compute: Drones and Robots

The bottleneck for autonomy has shifted from "How do we train the model?" to "How do we run it at the edge without a tether?"

• Defense: The Attritable Drone Swarm. In 2026, defense tech is defined by "compute-per-gram." We are seeing a move away from billion-dollar platforms toward disposable, autonomous drone swarms. These units don't just "fly," they use onboard edge inference to navigate EW-heavy (Electronic Warfare) environments where GPS and cloud links are severed. The edge is the mission.

• Consumer: The "Embodied" Assistant. 2025 was the year of the Humanoid pilot; 2026 is the year of the Humanoid product. Consumer robots are moving from "smart vacuums" to general-purpose helpers. This requires massive localized compute to handle vision-language-action models (VLA) in real-time. If your robot has to "check with the cloud" before catching a falling glass, it has already failed.

  2. Electricity at the Edge: Mobile Distributed Energy

You can’t have distributed compute without distributed power. The grid is struggling to keep up with the AI data center boom, forcing the edge to become energy-independent.

• Mobile Distributed Energy Resources (MDER): We are moving beyond stationary batteries. In 2026, we see the rise of "Energy on Wheels." Think of EV fleets not just as transport, but as mobile micro-nodes for the grid.

• The "Energy-Compute" Swap: We are seeing startups deploy modular, "drop-in" energy pods, combining high-density storage with edge servers. These pods act as a "buffer" for the local grid, charging when prices are low and providing both power and low-latency compute to local robotics fleets when demand peaks.

  3. The Great Displacement: From "Blue Collar" to "White Collar"

Perhaps my most provocative conviction for 2026 is the reversal of the automation narrative. While we spent decades fearing for the factory worker, the combination of Agentic AI and Kinetic Edge Compute is hitting the "Knowledge Worker" harder and faster.

• The Middle Management Collapse: Middle-tier roles in finance, legal, and operations (the "orchestrators" of information) are being displaced by autonomous agents. If your job is to move data from a PDF to a spreadsheet and then into a slide deck, an agentic workflow is now doing that for $0.01 per hour.

• The "Physical Premium": Curiously, 2026 is seeing a "Physical Premium." Because robotics (the "Hard Edge") is still more expensive to scale than software, tradespeople and on-site technicians are seeing wage growth, while entry-level white-collar roles are facing a "hiring freeze" that looks more like a permanent structural shift.

Robert’s Predictions

1. AI Will Cease to Exist as a Product Category and Becomes Infrastructure

By 2026, the most valuable AI companies will no longer market themselves as AI companies at all. The chatbox era fades as intelligence dissolves into products that simply work. Support resolves itself, workflows adapt automatically, and software anticipates intent instead of waiting for prompts. AI shifts from being a feature to being assumed infrastructure, like cloud computing or databases. The competitive edge moves away from model quality alone toward integration, reliability, and distribution. Companies still selling “AI-powered” features will start to look like those still advertising “cloud-based” software a decade ago.

2. The Open-Weight AI Ecosystem Becomes Strategically Unavoidable

In 2026, open-weight models stop being a research curiosity and become a board-level decision. Not because they beat closed models on benchmarks, but because they offer something more valuable: control. Enterprises and governments quietly adopt open models for internal systems, vision pipelines, and edge deployments where cost, customization, and sovereignty at times matter more than raw capability. This doesn’t play out as an ideological victory for open source, but as a pragmatic shift in the AI stack. Model labs lose monopoly leverage as value migrates toward ownership of workflows rather than access to intelligence.

Colin’s Predictions

1. The Return to In Person

We’re in the middle of a loneliness epidemic. Life keeps moving further online, and with the rise of AI generated content, deepfakes, and endless AI slop, the fatigue is going to be real (if not already). Think Zoom fatigue during COVID on steroids.

My bet is people start shifting their behavior in response. More “third spaces,” more social clubs, more community, more reasons to be physically around other people again. We’ll keep getting more efficient, faster, and more technologically advanced, but as humans, that next layer of Maslow’s hierarchy of needs will not be an AI agent, even if AI changes how we interface with the world.

2. Marquee IPOs

I think 2026 has a real chance of delivering a set of “marquee” IPOs, the kind that break out beyond tech and finance circles. Names like SpaceX, Anduril, and Anthropic.

What’s different this time is that consumer finance has become mainstream. Platforms like Robinhood in the U.S. and Wealthsimple in Canada have made public markets feel accessible, so retail investors can actually participate and care. If these companies go public, I’d expect the headlines to travel well outside the usual bubble. Don’t be surprised if the average person hears about it through their legacy news platform and it dominates a full news cycle.

There’s also a political angle here too: an administration eager to flex their “deal-making” chops will want big, visible wins, and IPOs like these can become part of that story.